The NBA package offers multiple modules which can be use to retrieve information and statistics for the National Basketball Association, such as team names, season stats, game schedules, and boxscore metrics.
The Boxscore module can be used to grab information from a specific game. Metrics range from number of points scored to the number of free throws made, to the assist rate and much more. The Boxscore can be easily queried by passing a boxscore's URI on sports-reference.com which can be retrieved from the Schedule class (see Schedule module below for more information on retrieving game-specific information).
from sportsreference.nba.boxscore import Boxscore
game_data = Boxscore('201806080CLE')
print(game_data.away_points) # Prints 108
print(game_data.home_points) # Prints 85
df = game_data.dataframe # Returns a Pandas DataFrame of game metricsThe Boxscore module also contains a Boxscores class which searches for all games played on a particular day and returns a dictionary of matchups between all teams on the requested day. The dictionary includes the names and abbreviations for each matchup as well as the boxscore link if applicable.
from datetime import datetime
from sportsreference.nba.boxscore import Boxscores
games_today = Boxscores(datetime.today())
print(games_today.games) # Prints a dictionary of all matchups for todayThe Boxscores class also allows the ability to query over a range of dates using a second optional parameter during instantiation of the class. To query a range of dates, enter the start date as the first parameter and the inclusive end date as the second parameter.
from datetime import datetime
from sportsreference.nba.boxscore import Boxscores
# Pulls all games between and including January 1, 2018 and January 5, 2018
games = Boxscores(datetime(2018, 1, 1), datetime(2018, 1, 5))
# Prints a dictionary of all results from January 1, 2018 and January 5,
# 2018
print(games.games)sportsreference.nba.boxscore
The Roster module contains detailed player information, allowing each player to be queried by their player ID using the Player class which has detailed information ranging from career points totals to single-season stats and player height, weight, and nationality. The following is an example on collecting career information for James Harden:
from sportsreference.nba.roster import Player
james_harden = Player('hardeja01')
print(james_harden.name) # Prints 'James Harden'
print(james_harden.points) # Prints Harden's career points total
# Prints a Pandas DataFrame of all relevant Harden stats per season
print(james_harden.dataframe)
print(james_harden.salary) # Prints Harden's career earnings
print(james_harden.contract) # Prints Harden's contract by yearly wagesBy default, the player's career stats are returned whenever a property is called. To get stats for a specific season, call the class instance with the season string. All future property requests will return the season-specific stats.
from sportsreference.nba.roster import Player
james_harden = Player('hardeja01') # Currently pulling career stats
print(james_harden.points) # Prints Harden's CAREER points total
# Prints Harden's points total only for the 2017-18 season.
print(james_harden('2017-18').points)
# Prints the number of games Harden played in the 2017-18 season.
print(james_harden.games_played)After requesting single-season stats, the career stats can be requested again by calling the class without arguments or with the 'Career' string passed.
from sportsreference.nba.roster import Player
james_harden = Player('hardeja01') # Currently pulling career stats
# Prints Harden's points total only for the 2017-18 season.
print(james_harden('2017-18').points)
print(james_harden('Career').points) # Prints Harden's career points totalIn addition, the Roster module also contains the Roster class which can be used to pull all players on a team's roster during a given season and creates instances of the Player class for each team member and adds them to a list to be easily queried.
from sportsreference.nba.roster import Roster
houston = Roster('HOU')
for player in houston.players:
# Prints the name of all players who played for Houston in the most
# recent season.
print(player.name)sportsreference.nba.roster
The Schedule module can be used to iterate over all games in a team's schedule to get game information such as the date, score, result, and more. Each game also has a link to the Boxscore class which has much more detailed information on the game metrics.
from sportsreference.nba.schedule import Schedule
houston_schedule = Schedule('HOU')
for game in houston_schedule:
print(game.date) # Prints the date the game was played
print(game.result) # Prints whether the team won or lost
# Creates an instance of the Boxscore class for the game.
boxscore = game.boxscoresportsreference.nba.schedule
The Teams module exposes information for all NBA teams including the team name and abbreviation, the number of games they won during the season, the total number of shots they've blocked, and much more.
from sportsreference.nba.teams import Teams
teams = Teams()
for team in teams:
print(team.name) # Prints the team's name
print(team.blocks) # Prints the team's total blocked shotsEach Team instance contains a link to the Schedule class which enables easy iteration over all games for a particular team. A Pandas DataFrame can also be queried to easily grab all stats for all games.
from sportsreference.nba.teams import Teams
teams = Teams()
for team in teams:
schedule = team.schedule # Returns a Schedule instance for each team
# Returns a Pandas DataFrame of all metrics for all game Boxscores for
# a season.
df = team.schedule.dataframe_extendedLastly, each Team instance also contains a link to the Roster class which enables players from the team to be easily queried. Each Roster instance contains detailed stats and information for each player on the team.
from sportsreference.nba.teams import Teams
teams = Teams()
for team in teams:
# Creates an instance of the roster class for each player on the team.
roster = team.roster
for player in roster.players:
print(player.name) # Prints the name of each player on the team.sportsreference.nba.teams